Netlfix Insights and Analysis

Loading the dataset

import pandas as pd
import warnings
warnings.filterwarnings("ignore")
from wordcloud import WordCloud, STOPWORDS
import seaborn as sns
import matplotlib.pyplot as plt

netflix_df = pd.read_csv("netflix_titles.csv")
netflix_df.head(10)
netflix_df.to_excel('nefflix_dataset.xlsx') # I did this so that I could view it in an excel file as I enjoy looking at my data while analyzing it

netflix_df.describe()

	release_year
count	8807.000000
mean	2014.180198
std	8.819312
min	1925.000000
25%	2013.000000
50%	2017.000000
75%	2019.000000
max	2021.000000

netflix_df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 8807 entries, 0 to 8806
Data columns (total 12 columns):
 #   Column        Non-Null Count  Dtype 
---  ------        --------------  ----- 
 0   show_id       8807 non-null   object
 1   type          8807 non-null   object
 2   title         8807 non-null   object
 3   director      6173 non-null   object
 4   cast          7982 non-null   object
 5   country       7976 non-null   object
 6   date_added    8797 non-null   object
 7   release_year  8807 non-null   int64 
 8   rating        8803 non-null   object
 9   duration      8804 non-null   object
 10  listed_in     8807 non-null   object
 11  description   8807 non-null   object
dtypes: int64(1), object(11)
memory usage: 825.8+ KB

Transformaing the dataset for analysis

netflix_df['date_added'] = pd.to_datetime(netflix_df['date_added'], format='mixed')

netflix_shows=netflix_df[netflix_df['type']=='TV Show']
netflix_shows['show_id'] = netflix_shows['show_id'].str.replace('s','')
netflix_movies=netflix_df[netflix_df['type']=='Movie']
netflix_movies['show_id'] = netflix_movies['show_id'].str.replace('s','')

netflix_movies.head(5)

	show_id	type	title	director	cast	country	date_added	release_year	rating	duration	listed_in	description
0	1	Movie	Dick Johnson Is Dead	Kirsten Johnson	NaN	United States	2021-09-25	2020	PG-13	90 min	Documentaries	As her father nears the end of his life, filmm...
6	7	Movie	My Little Pony: A New Generation	Robert Cullen, José Luis Ucha	Vanessa Hudgens, Kimiko Glenn, James Marsden, ...	NaN	2021-09-24	2021	PG	91 min	Children & Family Movies	Equestria's divided. But a bright-eyed hero be...
7	8	Movie	Sankofa	Haile Gerima	Kofi Ghanaba, Oyafunmike Ogunlano, Alexandra D...	United States, Ghana, Burkina Faso, United Kin...	2021-09-24	1993	TV-MA	125 min	Dramas, Independent Movies, International Movies	On a photo shoot in Ghana, an American model s...
9	10	Movie	The Starling	Theodore Melfi	Melissa McCarthy, Chris O'Dowd, Kevin Kline, T...	United States	2021-09-24	2021	PG-13	104 min	Comedies, Dramas	A woman adjusting to life after a loss contend...
12	13	Movie	Je Suis Karl	Christian Schwochow	Luna Wedler, Jannis Niewöhner, Milan Peschel, ...	Germany, Czech Republic	2021-09-23	2021	TV-MA	127 min	Dramas, International Movies	After most of her family is murdered in a terr...

netflix_shows.head(5)

	show_id	type	title	director	cast	country	date_added	release_year	rating	duration	listed_in	description
1	2	TV Show	Blood & Water	NaN	Ama Qamata, Khosi Ngema, Gail Mabalane, Thaban...	South Africa	2021-09-24	2021	TV-MA	2 Seasons	International TV Shows, TV Dramas, TV Mysteries	After crossing paths at a party, a Cape Town t...
2	3	TV Show	Ganglands	Julien Leclercq	Sami Bouajila, Tracy Gotoas, Samuel Jouy, Nabi...	NaN	2021-09-24	2021	TV-MA	1 Season	Crime TV Shows, International TV Shows, TV Act...	To protect his family from a powerful drug lor...
3	4	TV Show	Jailbirds New Orleans	NaN	NaN	NaN	2021-09-24	2021	TV-MA	1 Season	Docuseries, Reality TV	Feuds, flirtations and toilet talk go down amo...
4	5	TV Show	Kota Factory	NaN	Mayur More, Jitendra Kumar, Ranjan Raj, Alam K...	India	2021-09-24	2021	TV-MA	2 Seasons	International TV Shows, Romantic TV Shows, TV ...	In a city of coaching centers known to train I...
5	6	TV Show	Midnight Mass	Mike Flanagan	Kate Siegel, Zach Gilford, Hamish Linklater, H...	NaN	2021-09-24	2021	TV-MA	1 Season	TV Dramas, TV Horror, TV Mysteries	The arrival of a charismatic young priest brin...

directors_movies = netflix_movies['director'].unique()
directors_shows = netflix_shows['director'].unique()
movies_categories = netflix_movies['listed_in'].unique()
shows_categories = netflix_shows['listed_in'].unique()

Inital Analysis

lets first inspect how many tv shows and movies are there in out dataset

category_counts = netflix_df['type'].value_counts()
type_countplot = sns.countplot(data=netflix_df, x='type')
for i, count in enumerate(category_counts):
    plt.text(i, count, str(count), ha='center', va='bottom')

lets inspect how many movies are added to netflix every year

netflix_movies['year_added'] = netflix_movies['date_added'].dt.year
year_counts = netflix_movies['year_added'].value_counts()
year_counts = year_counts.sort_index()
plt.figure(figsize=(10,5))
sns.countplot(data=netflix_movies, x='year_added')
for i, count in enumerate(year_counts):
    plt.text(i, count, str(count), ha='center', va='bottom')
plt.show()

lets see how many series are added into the netflix platform every year

netflix_shows['year_added'] = netflix_shows['date_added'].dt.year
year_counts = netflix_shows['year_added'].value_counts()
year_counts = year_counts.sort_index()
plt.figure(figsize=(10,5))
sns.countplot(data=netflix_shows, x='year_added')
for i, count in enumerate(year_counts):
    plt.text(i, count, str(count), ha='center', va='bottom')
plt.show()

It is seen that from the year of 2016 the amount of movies and series added on to the platform increasing. But it cannot be ignored that from the year 2020 there is a drop in both series added and moveis added. This could be a result of covid pandemic or a certain finanical issue.

Lets look at the vastness of release years of these movies and series to understand which period of movies are prominent in the platform

year_counts = netflix_shows['release_year'].value_counts()
year_counts = year_counts.sort_index()
plt.figure(figsize=(32,15))
plt.title("Histogram showcasing the release dates of series added on the netflix platfor ( Shows)")
sns.countplot(data=netflix_shows, x='release_year')
for i, count in enumerate(year_counts):
    plt.text(i, count, str(count), ha='center', va='bottom')
plt.show()

year_counts = netflix_movies['release_year'].value_counts()
year_counts = year_counts.sort_index()
plt.figure(figsize=(35,15))
plt.title("Histogram showcasing the release dates of movies added on the netflix platform (Movies)")
sns.countplot(data=netflix_movies, x='release_year')
for i, count in enumerate(year_counts):
    plt.text(i, count, str(count), ha='center', va='bottom')
plt.show()

when we look at the movies and series added onto the platform, the period from 2011 to 2019 there was a explonention increase in the addition of movies and series in that period. With respect to movies, the trend continued to the year 2020 while in series, there is a decline in the amount of series added from the period of 2019-2021.

Lets see amoungst all these shows and movies what is the division on the ratings

rating_counts = netflix_movies['rating'].value_counts()
plt.figure(figsize=(20,5))
plt.title("Histogram showcasing the rating division amongst the movies")
order = rating_counts.index
sns.countplot(data=netflix_movies, x='rating', order=order)
for i, count in enumerate(rating_counts):
    plt.text(i, count, str(count), ha='center', va='bottom')
plt.show()

rating_counts = netflix_shows['rating'].value_counts()
plt.figure(figsize=(20,5))
plt.title("Histogram showcasing the rating division amongst the shows")
order = rating_counts.index
sns.countplot(data=netflix_shows, x='rating', order=order)
for i, count in enumerate(rating_counts):
    plt.text(i, count, str(count), ha='center', va='bottom')
plt.show()

From the above 2 graph we can see that TV-MA is the most popular group rating amoungs both tv shows and movies which counts for almost 3207 movies and shows combined. Similarily the top 3 ratings are TV-MA, TV-14 and TV-PG for shows and R for movies.

lets now see the duration analysis of our movies and shows

#creating bins for the duration of movies
netflix_movies['duration'] = netflix_movies['duration'].str.extract('(\d+)', expand=False).fillna(0).astype(int)
netflix_movies['duration_cat'] = pd.cut(netflix_movies['duration'], bins=[0,60,120, float('inf')], labels=['1 hour', '2 hour', '2+ hour'])

# ploting
cat_count = netflix_movies['duration_cat'].value_counts()
plt.figure(figsize=(10,5))
sns.countplot(data=netflix_movies, x='duration_cat', order=['1 hour', '2 hour', '2+ hour'])

for i, count in enumerate(cat_count):
    plt.text(i, count, str(count), ha='center', va='bottom')


plt.xlabel("duration category")
plt.ylabel("Number of Movies")
plt.title('Number of movies by duration')

plt.legend()
plt.show()

No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.

# ploting
cat_count = netflix_shows['duration'].value_counts()
order = cat_count.index
plt.figure(figsize=(20,5))
sns.countplot(data=netflix_shows, x='duration', order=order)

for i, count in enumerate(cat_count):
    plt.text(i, count, str(count), ha='center', va='bottom')


plt.xlabel("duration category")
plt.ylabel("Number of Movies")
plt.title('Number of shows divided with seasons')

plt.legend()
plt.show()

No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.

Lets examin the "Listed_in" column which is the column that represents where these movies/shows are showcased in the platform or what genre they are in

genre_count = pd.DataFrame(netflix_movies['listed_in'].str.split(',').explode().value_counts())
genre_count.reset_index(inplace=True)
genre_count.columns = ['category', 'count']

plt.figure(figsize=(10,10))
sns.barplot(data=genre_count, x='count', y='category')

plt.xlabel("count")
plt.ylabel("categories")
plt.title("count of movies in each genre ( A single movie can represent more than one genre)")
plt.show()

genre_count = pd.DataFrame(netflix_shows['listed_in'].str.split(',').explode().value_counts())
genre_count.reset_index(inplace=True)
genre_count.columns = ['category', 'count']

plt.figure(figsize=(10,10))
sns.barplot(data=genre_count, x='count', y='category')

plt.xlabel("count")
plt.ylabel("categories")
plt.title("count of movies in each genre ( A single show can represent more than one genre)")
plt.show()

Let us now examine the country wise production/addition of movies and series in Netflix platform

country_count_movies = pd.DataFrame(netflix_movies['country'].str.split(',').explode().str.strip().value_counts())
country_count_movies .reset_index(inplace=True)
country_count_movies .columns = ['category', 'count']
country_count_movies = country_count_movies [:20]

plt.figure(figsize=(10,10))
sns.barplot(data=country_count_movies, x='count', y='category')

plt.xlabel("count")
plt.ylabel("Countries Production was done")
plt.title("countries where production of movies took place (top 20)")
plt.show()

country_count_shows = pd.DataFrame(netflix_shows['country'].str.split(',').explode().str.strip().value_counts())
country_count_shows.reset_index(inplace=True)
country_count_shows.columns = ['category', 'count']
country_count_shows = country_count_shows[:20]

plt.figure(figsize=(10,10))
sns.barplot(data=country_count_shows, x='count', y='category')

plt.xlabel("count")
plt.ylabel("Countries Production was done")
plt.title("countries where production of shows took place")
plt.show()

Lets examine the directors who makes movies and tv shows

director_count = pd.DataFrame(netflix_movies['director'].str.split(',').explode().value_counts())
director_count.reset_index(inplace=True)
director_count.columns = ['cast', 'count']
director_count = director_count[:20]

plt.figure(figsize=(10,10))
sns.barplot(data=director_count, x='count', y='cast')

plt.xlabel("count")
plt.ylabel("cast")
plt.title("Directors who make the most movies")
plt.show()

director_count = pd.DataFrame(netflix_shows['director'].str.split(',').explode().value_counts())
director_count.reset_index(inplace=True)
director_count.columns = ['cast', 'count']
director_count = director_count[:20]

plt.figure(figsize=(10,10))
sns.barplot(data=director_count, x='count', y='cast')

plt.xlabel("count")
plt.ylabel("cast")
plt.title("Directors who make the most shows")
plt.show()

lets examine the top casts the produce them most amount of movies that appear in netflix

cast_count = pd.DataFrame(netflix_movies['cast'].str.split(',').explode().value_counts())
cast_count.reset_index(inplace=True)
cast_count.columns = ['cast', 'count']
cast_count = cast_count[:25]
#cast_count['cast'] = cast_count['cast'].str.replace(r'\$', 's')
#cast_counts = cast_count[~cast_count['cast'].str.contains('[^a-zA-Z\s]', regex=True)]
cast_count['cast'] = cast_count['cast'].str.replace('Joey Bada$$', 'Joey Badass')

plt.figure(figsize=(10,10))
sns.barplot(data=cast_count, x='count', y='cast')

plt.xlabel("count")
plt.ylabel("cast")
plt.title("Casts who make the most movies featured on netflix")
plt.show()

cast_count = pd.DataFrame(netflix_shows['cast'].str.split(',').explode().value_counts())
cast_count.reset_index(inplace=True)
cast_count.columns = ['cast', 'count']
cast_count = cast_count[:25]
#cast_count['cast'] = cast_count['cast'].str.replace(r'\$', 's')
#cast_counts = cast_count[~cast_count['cast'].str.contains('[^a-zA-Z\s]', regex=True)]
cast_count['cast'] = cast_count['cast'].str.replace('Joey Bada$$', 'Joey Badass')

plt.figure(figsize=(10,10))
sns.barplot(data=cast_count, x='count', y='cast')

plt.xlabel("count")
plt.ylabel("cast")
plt.title("Casts who make the most shows featured on netflix")
plt.show()

Indept Analysis of the Dataset

here we will try and make assumption of the data and see which of the assuptions are correct

Which is the best month to release the movie or series on netflix? ( we need to make sure the movie/series has less compitetion and hence the month with the least amount of releases)

netflix_movies['added month'] = netflix_movies['date_added'].dt.month
netflix_movies['added year'] = netflix_movies['date_added'].dt.year
# Create a pivot table to count the number of movies by year and month
movies_count = netflix_movies.pivot_table(index='added year', columns='added month', aggfunc='size', fill_value=0)

# Create the heatmap
plt.figure(figsize=(12, 6))
sns.heatmap(movies_count, annot=True, cmap='YlGnBu', fmt='d')

# Set labels and title
plt.xlabel('Month')
plt.ylabel('Year')
plt.title('Number of Movies Produced by Month and Year')

# Show the plot
plt.show()

when we look at the above table it can be seen that the best time to release a movie on netflix such that there isnt much competition would be the months of Feburary, May, August and September.

netflix_shows['added month'] = netflix_shows['date_added'].dt.month
netflix_shows['added year'] = netflix_shows['date_added'].dt.year
# Create a pivot table to count the number of movies by year and month
shows_count = netflix_shows.pivot_table(index='added year', columns='added month', aggfunc='size', fill_value=0)

# Create the heatmap
plt.figure(figsize=(12, 6))
sns.heatmap(shows_count, annot=True, cmap='YlGnBu', fmt='d')

# Set labels and title
plt.xlabel('Month')
plt.ylabel('Year')
plt.title('Number of shows Produced by Month and Year')

# Show the plot
plt.show()

When looking at the best time to release a tv series on netflix, it is seen that it has been pretty evenly spread out and hence would not make much of a different. If we isolate only 2021, then definitly the first half of the eyar would be the best bet in terms of performance.

Country Wise Analysis

Lets now dig a little deeper into the top 5 countrys and find out the following....

1. Which are the top genres avaliable to each of th countries
2. Making a word cloud of the top things that are seen common storyline amongst each country

top_five_countries_movies = country_count_movies['category'][:5].tolist()
print("top 5 countries that produce movies",top_five_countries_movies)
top_five_countries_shows = country_count_shows['category'][:5].tolist()
print("top countries that produce tv shows",top_five_countries_shows)

top 5 countries that produce movies ['United States', 'India', 'United Kingdom', 'Canada', 'France']
top countries that produce tv shows ['United States', 'United Kingdom', 'Japan', 'South Korea', 'Canada']

Contribution of top 5 countries in each genre (Movies)

# creating a new dataframe with only the needed columns
df_filtered = netflix_movies[['listed_in', 'country', 'title']]
df_filtered['listed_in'] = df_filtered['listed_in'].str.split(', ')
df_filtered = df_filtered.explode('listed_in')
df_filtered['country'] = df_filtered['country'].str.split(', ')
df_filtered = df_filtered.explode('country')
df_filtered = df_filtered[df_filtered['country'].isin(top_five_countries_movies)]
genre_country_counts = df_filtered.groupby(['listed_in', 'country']).size().reset_index(name='count')


# Create a bar chart
plt.figure(figsize=(12, 16))
sns.barplot(data=genre_country_counts, x='count', y='listed_in', hue='country')

# Set labels and title
plt.xlabel('Count')
plt.ylabel('Genre')
plt.title('Number of Titles Produced by Genre and Country(Movies)')

# Show the plot
plt.show()

Contribution of top 5 countries in each genre (shows)

# creating a new dataframe with only the needed columns
df_filtered = netflix_shows[['listed_in', 'country', 'title']]
df_filtered['listed_in'] = df_filtered['listed_in'].str.split(', ')
df_filtered = df_filtered.explode('listed_in')
df_filtered['country'] = df_filtered['country'].str.split(', ')
df_filtered = df_filtered.explode('country')
df_filtered = df_filtered[df_filtered['country'].isin(top_five_countries_shows)]
genre_country_counts = df_filtered.groupby(['listed_in', 'country']).size().reset_index(name='count')


# Create a bar chart
plt.figure(figsize=(12, 16))
sns.barplot(data=genre_country_counts, x='count', y='listed_in', hue='country')

# Set labels and title
plt.xlabel('Count')
plt.ylabel('Genre')
plt.title('Number of Titles Produced by Genre and Country (Shows)')

# Show the plot
plt.show()

Word cloud of the the top 25 words seen in the description amongst the top 5 countries

df_filtered = netflix_movies[netflix_movies['country'].isin(top_five_countries_movies)]
for country in top_five_countries_movies:
    country_descriptions = df_filtered[df_filtered['country'] == country]['description']
    all_descriptions = ' '.join(country_descriptions)
    # Remove stop words from the descriptions
    stopwords = set(STOPWORDS)
    stopwords.update(["movie", "film", "story", "show"])  # Add custom stop words if needed
    # Create a WordCloud object with cleaned descriptions
    wordcloud = WordCloud(width=800, height=400, max_words=25, background_color='white', stopwords=stopwords).generate(all_descriptions)
    
    # Display the word cloud
    plt.figure(figsize=(12, 6))
    plt.imshow(wordcloud, interpolation='bilinear')
    plt.axis('off')
    plt.title(f"Top 25 Common Words in Movie Descriptions for {country}")
    plt.show()

    

from wordcloud import WordCloud, STOPWORDS
df_filtered = netflix_shows[netflix_shows['country'].isin(top_five_countries_shows)]
for country in top_five_countries_shows:
    country_descriptions = df_filtered[df_filtered['country'] == country]['description']
    all_descriptions = ' '.join(country_descriptions)
    # Remove stop words from the descriptions
    stopwords = set(STOPWORDS)
    stopwords.update(["movie", "film", "story", "show"])  # Add custom stop words if needed
    # Create a WordCloud object with cleaned descriptions
    wordcloud = WordCloud(width=800, height=400, max_words=25, background_color='white', stopwords=stopwords).generate(all_descriptions)
    
    # Display the word cloud
    plt.figure(figsize=(12, 6))
    plt.imshow(wordcloud, interpolation='bilinear')
    plt.axis('off')
    plt.title(f"Top 25 Common Words in show descriptions for {country}")
    plt.show()


    

This in-depth analysis of Netflix titles has been compiled and executed by Rohit Sanjay.